Instructions for use

Title Comparison of the nephroprotective effects of non-steroidal anti-inflammatory drugs on cisplatin-inducednephrotoxicity in vitro and in vivo

Author(s) Okamoto, Keisuke; Saito, Yoshitaka; Narumi, Katsuya; Furugen, Ayako; Iseki, Ken; Kobayashi, Masaki

Citation European journal of pharmacology, 884, 173339https://doi.org/10.1016/j.ejphar.2020.173339

Issue Date 2020-10-05

Doc URL http://hdl.handle.net/2115/82868

Rights ©2020. This manuscript version is made available under the CC-BY-NC-ND 4.0 licensehttp://creativecommons.org/licenses/by-nc-nd/4.0/

Rights(URL) http://creativecommons.org/licenses/by-nc-nd/4.0/

Type article (author version)

File Information WoS_95814_Kobayashi.pdf

Hokkaido University Collection of Scholarly and Academic Papers : HUSCAP

1

Title

Comparison of the nephroprotective effects of non-steroidal anti-inflammatory drugs on

cisplatin-induced nephrotoxicity in vitro and in vivo

Keisuke Okamoto a, Yoshitaka Saito b, Katsuya Narumi a, Ayako Furugen a, Ken Iseki a,

Masaki Kobayashi a

a Laboratory of Clinical Pharmaceutics & Therapeutics, Division of Pharmasciences, Faculty

of Pharmaceutical Sciences, Hokkaido University, Kita-12-jo, Nishi-6-chome, Kita-ku,

Sapporo, 060-0812, Japan

b Department of Pharmacy, Hokkaido University Hospital, Kita-14-jo, Nishi-5-chome,

Kita-ku, Sapporo, 060-8648, Japan

Corresponding Author: Masaki Kobayashi

Laboratory of Clinical Pharmaceutics & Therapeutics, Division of Pharmasciences, Faculty of

Pharmaceutical Sciences, Hokkaido University, Kita-12-jo, Nishi-6-chome, Kita-ku, Sapporo,

060-0812, Japan

E-mail: masaki@pharm.hokudai.ac.jp

2

Abstract

Cisplatin (CDDP) is an anticancer drug, often used in the treatment of several types

of cancers. CDDP-induced nephrotoxicity (CIN) is one of the most severe adverse events

associated with the use of CDDP. It has been suggested that the co-administration of

non-steroidal anti-inflammatory drugs (NSAIDs) is a risk factor for CIN. However, the

specific NSAIDs that affect CIN and the precise mechanisms underlying this interaction

remain unclear. Hence, we aimed to evaluate the effect of NSAIDs on CDDP-induced

cytotoxicity in vitro and confirmed the results in vivo. Using the epithelioid clone of the

normal rat kidney cells (NRK-52E cells), we assessed the effects of 17 NSAIDs on

CDDP-induced cytotoxicity all at once using the MTT assay. Furthermore, we evaluated two

NSAIDs, which significantly attenuated or enhanced CDDP-induced cytotoxicity, in vivo.

Wistar rats were treated with CDDP (5 mg/kg, i.p., day 1) and NSAIDs (p.o., day 1–4), and

the kidneys were excised on day 5. Our results demonstrated that several NSAIDs attenuated,

while others enhanced CDDP-induced cytotoxicity. Celecoxib significantly attenuated and

flurbiprofen markedly enhanced cell dysfunction by CDDP. These results were reproduced in

vivo as celecoxib decreased and flurbiprofen increased the expression of kidney injury

molecule 1 (Kim-1) mRNA, a sensitive kidney injury marker, compared to the CDDP group.

Moreover, celecoxib increased the antioxidant and autophagy markers quantified by qPCR in

vitro and prevented a decrease in body weight induced by CDDP in vivo. In conclusion, we

3

revealed that celecoxib significantly attenuated CIN in vitro and in vivo.

Keywords

Cisplatin; Nephrotoxicity; NSAIDs; Celecoxib

1. Introduction

Cisplatin (cis-dichloro-diammine platinum, CDDP) is widely used as an anticancer

agent in many types of cancer, including lung, gastric, and head and neck cancer (Rosenberg

et al., 1969). CDDP-induced nephrotoxicity (CIN) is one of the most serious adverse effects

caused by CDDP (Prestayko et al., 1979), and it is dose-dependent, cumulative, and usually

reversible (Miller et al., 2010; Pabla and Dong, 2008). CIN occurs in 30–40% of patients

administered CDDP (Yoshida et al., 2014). CIN mainly affects the S3 segment of the

proximal tubule located in the outer medulla, and the thick ascending limb of the loop of

Henle (Dobyan et al., 1980). The suggested mechanisms of CIN include oxidative stress,

mitochondrial dysfunction, DNA damage, increased tumor necrosis factor, and inhibition of

protein synthesis (Kawai et al., 2006; Park et al., 2002; Tsuruya et al., 2003). In spite of

adopting hydration as an approach to prevent CIN, it fails to sufficiently attenuate CIN (de

Jongh et al., 2003).

In a retrospective clinical study, our research group demonstrated that the

4

co-administration of non-steroidal anti-inflammatory drugs (NSAIDs) enhances CIN (Saito et

al., 2017a). NSAIDs demonstrate their anti-inflammatory effect through the inhibition of

cyclooxygenase (COX) activity (Vane, 1971). There are two isoforms of COX, COX-1 is

constantly expressed and COX-2 is expressed in response to inflammatory signals (Loftin et

al., 2002). Some NSAIDs inhibit COX non-selectively, while others inhibit COX-2 selectively

(Warner et al., 1999). While the strong inhibition of COX-1 causes gastrointestinal

dysfunction, selective COX-2 inhibitors keep the stomach protected by targeting only COX-2.

The World Health Organization (WHO) recommends the administration of NSAIDs for

cancer pain (World Health Organization, 1996). Hence, NSAIDs and anticancer drugs are

often combined in clinical use.

Together with fellow researchers, we have reported that the co-administration of

NSAIDs is a risk factor for CIN (Kidera et al., 2014; Sato et al., 2016). However, these results

were established based on retrospective clinical studies and secondary assessments. Moreover,

the specific NSAIDs that affect CIN and the precise mechanism underlying the possible

interaction remain unclear. Although several reports have assessed the relationship between

CDDP and some NSAIDs in vitro and in vivo (Fernández-Martínez et al., 2016; Honma et al.,

2013), there is no report comparing multiple NSAIDs under the same condition. In this study,

we aimed to assess the effect of NSAIDs on CDDP-induced cytotoxicity all at once in vitro

and to reproduce these effects in vivo by using two NSAIDs which significantly attenuated or

5

enhanced cell injury induced in vitro by CDDP, to determine the appropriate

co-administration in patients administered CDDP.

2. Materials and Methods

2.1. Chemicals

CDDP was purchased from FUJIFILM Wako Pure Chemical Corp. (Osaka, Japan).

Celecoxib was purchased from the Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan).

Flurbiprofen was purchased from Sigma Aldrich (St. Louis, MO, USA). All other chemicals

and reagents were commercially available and of the highest purity possible. NSAIDs were

chosen for the following reasons: (1) having a dosage form for oral administration; (2) they

were not a prodrug; (3) easily available.

2.2. Cell culture

The epithelioid clone of normal rat kidney cells (NRK-52E cells, JCRB Cell Bank,

Osaka, Japan) was cultured in DMEM (Sigma Aldrich, St. Louis, MO, USA), supplemented

with 10% fetal bovine serum and 1% penicillin-streptomycin (Sigma Aldrich, St. Louis, MO,

USA) in a humidified atmosphere of 5% CO2 at 37°C. NRK-52E cells with a passage number

of 25–40 were used in the experiment.

2.3. Animals and experimental design

Male Wistar rats (7 weeks old) were procured from JLA (Tokyo, Japan). All rats were

6

housed in an animal maintenance facility, in a room with controlled temperature (23°C) and

moisture (60 ± 10%) conditions and a 12 h light–dark cycle. All rats were allowed free access

to demineralized diet pellets and water. All animal experiments were conducted in accordance

with the guidelines for the Care and Use of Laboratory Animals of Hokkaido University, and

all experimental protocols were reviewed and approved by the Hokkaido University Animal

Care Committee in accordance with the “Guide for the Care and Use of Laboratory Animals”

(approval number: 17-0005).

Rats were treated with CDDP (5 mg/kg, 2 mg/ml in saline, i.p., day 1) or saline, and

celecoxib (30 mg/kg/day, 30 mg/ml in methyl cellulose, p.o., day 1–4) or flurbiprofen (10

mg/kg/day, 10 mg/ml in methyl cellulose, p.o., day 1–4), or methyl cellulose. Rats were

divided into six groups: (1) Control group, saline and methyl cellulose; (2) CDDP group,

CDDP and methyl cellulose; (3) CDDP + celecoxib group; (4) CDDP + flurbiprofen group;

(5) celecoxib group; (6) flurbiprofen group. On day 5, blood samples (200–300 μl) were

collected from the tail vein. After blood collection, the rats were anesthetized with sevoflurane,

euthanized, and the kidneys were immediately excised. The kidney tissues were washed in

saline and stored at −80°C until further analysis.

2.4. Measurement of mRNA expression

Total RNA was extracted using an ISOGEN II kit (Nippon Gene, Tokyo, Japan),

according to the manufacturer’s protocol. The cortex from the kidney was then sliced and

7

total RNA was extracted. Total RNA was used to prepare complementary DNA by reverse

transcription using ReverTra Ace (Toyobo, Osaka, Japan). The level of mRNA was measured

by RT-PCR or qPCR. Supplemental Table 1 shows the primer sequences used for the PCR

amplification.

RT-PCR was performed using a KAPATaq Extra kit (NIPPON Genetics, Tokyo,

Japan) and T100TM Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, USA). The

RT-PCR thermocycling protocol was: 30 cycles of denaturation at 95°C for 30 s; annealing at

56.5°C–63.5°C for 30 s; and extension at 72°C for 30 s. Products were size-fractionated on

1.2 % agarose gel and stained using ethidium bromide. The band was detected by using

LAS-1000 (GE Healthcare UK Ltd., Buckinghamshire, UK) and band intensities were

analyzed by using ImageJ analysis software (NIH, Bethesda, MD, USA). qPCR was

performed by using a KAPA SYBR Fast qPCR kit (Kapa Biosystems, Wilmington, MA,

USA) and Mx3000p (Agilent Technologies, Santa Clara, CA, USA). The qPCR

thermocycling protocol was: 40 cycles of denaturation at 95°C for 5 s; annealing at

58°C–60°C for 20 s; and extension at 72°C for 30 s. Standard curves were constructed for

each target and housekeeping gene. The software calculated the relative amount of the target

gene and the housekeeping gene based on the threshold cycles.

2.5. Cell viability assay

The cells were separately seeded in 96-well plates at a density of 5,000 cells/well in

8

the culture medium. After 24 h, the cells were treated with CDDP and/or some reagents

(containing 0.1% dimethyl sulfoxide (DMSO)) for 48 h, and cell viability was measured by

the MTT assay using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide in

accordance with the manufacturer’s instructions. The absorbance of the resulting reaction

solution was measured at 540 nm for samples and at 690 nm for the reference wavelength.

The concentration of 50% inhibition for cell viability (IC50) was calculated using SigmaPlot

14 (HULINKS Inc., Tokyo, Japan).

2.6. Superoxide scavenging assay

The superoxide scavenging ability was measured using

2-methyl-6-p-methoxyphenylethynyl-imidazopyrazinone (MPEC) (ATTO, Tokyo, Japan)

according to the manufacturer’s protocol. All reagents were dissolved in DMSO. The

concentrations of NSAIDs are presented in Table 1. Each reaction solution was placed in a

384-well white plate, and the chemiluminescence was measured.

2.7. Measurement of serum creatinine level

The rat blood samples were centrifuged at 1,200 × g for 20 min at 4°C. The serum

creatinine level in the supernatant was measured by using LabAssayTM Creatinine (FUJIFILM

Wako Pure Chemical Corp, Osaka, Japan) in accordance with the manufacturer’s protocol.

2.8. Histological examination

Kidney samples were fixed in 10% buffered formalin. The fixed tissue was

9

embedded in paraffin and sectioned. The prepared slides were stained with hematoxylin-eosin

and observed by using a BZ-9000 (KEYENCE, Osaka, Japan).

2.9. Statistical analysis

Statistical analyses of data were performed using the unpaired Student’s t-test, or

one-way ANOVA followed by Tukey’s post-hoc test, or one-way ANOVA followed by

Dunnett’s test or Pearson’s correlation coefficient. Data were analyzed using SigmaPlot 14,

and P < 0.05 was considered statistically significant.

3. Results

3.1. The characteristics of NRK-52E cells treated with CDDP

We compared renal transporter mRNA expression between NRK-52E cells and the

rat kidney cortex. Since most of renal transporters, including the organic cation transporter 2

(Oct2), copper transporter (Ctr1) and multidrug and toxin extrusion 1 (Mate1) mainly

transporting CDDP (Yokoo et al., 2009), were expressed in the NRK-52E cells, we thought

that this cell line reflected the rat kidney function (Fig. 1A). Following exposure of CDDP to

NRK-52E cells for 48 h, the IC50 was calculated as 7.4 ± 0.2 µM (Fig. 1B). Moreover, the

mRNA levels of Clusterin, a kidney injury marker, were significantly increased following

treatment with 10 µM CDDP (Fig. 1C).

3.2. The effects of 17 NSAIDs on CDDP-induced cytotoxicity in screening

10

Cell viability following exposure to NSAIDs for 48 h is shown in Table 1. IC50

values could not be calculated for most NSAIDs, except celecoxib. Next, the ratios of IC50 of

co-addition of CDDP and NSAIDs to IC50 of CDDP alone in the MTT assay screening are

shown in Fig. 2A, and the concentration of NSAIDs was established as that which did not

injure the NRK-52E cells within a range of 95–120% cell viability. Hence, among the 17

NSAIDs screened, it was clearly demonstrated that celecoxib significantly attenuated and

flurbiprofen markedly enhanced CDDP-induced cytotoxicity. Moreover, the co-addition of

celecoxib significantly decreased and flurbiprofen co-addition increased the Clusterin mRNA

level compared to CDDP alone (Fig. 2B and Fig. 2C).

3.3. Celecoxib attenuated CDDP-induced cytotoxicity via antioxidant effects.

Rashtchizadeh et al. reported that CIN is prevented by antioxidant effects

(Rashtchizadeh et al., 2019). Hence, we focused on antioxidant markers. Celecoxib

co-addition significantly increased the mRNA levels of the antioxidant markers including

heme oxygenase 1 (Ho-1, Fig. 3A), superoxide dismutase 1 (Sod1, Fig. 3B) and

NF-E2-related factor 2 (Nrf2, Fig. 3C). Moreover, we observed correlations between the IC50

of cell viability and the antioxidant markers (Table 2). To elucidate the mechanism of

celecoxib mediated oxidative stress reduction, we examined the superoxide scavenging ability

using MPEC (Kobayashi and Kanai, 2013). However, NSAIDs failed to demonstrate

superoxide scavenging ability themselves (Supplemental Fig. 1). Therefore, celecoxib

11

attenuated CDDP-induced cytotoxicity by enhancing the resistance to oxidative stress in cells,

and not by directly reducing oxidative stress through superoxide scavenging.

3.4. Autophagy was related to celecoxib attenuation of CDDP-induced cytotoxicity.

It has been suggested that attenuating CIN is related not only to antioxidant effects

but also autophagy (Kaushal and Shah, 2016). Thus, we also evaluated autophagy markers.

The changes in mRNA levels of autophagy markers, autophagy related gene (Atg) 5, Atg7, Lc

3, and Beclin 1, are shown in Fig. 4A. Celecoxib co-addition increased Atg5, Atg7, and Lc 3

mRNA levels in comparison to CDDP alone. Additionally, on evaluating the changes in Atg5

mRNA level, in each of the 17 NSAIDs co-additions, we observed that celecoxib significantly

increased the Atg5 mRNA level (Fig. 4B). Moreover, a correlation was observed between

Atg5 mRNA level and cell viability like antioxidant effects (Table 2). Hence, it was suggested

that autophagy, as well as antioxidant effects, could be responsible for the attenuation of

CDDP-induced cytotoxicity mediated by celecoxib.

3.5. The effects of celecoxib on CIN in vivo

The changes in the body weight of rats administered CDDP, with and without

celecoxib and flurbiprofen, are shown in Table 3. In our study, CDDP-treated rats

demonstrated a decreased body weight. However, celecoxib co-administration increased the

body weight to the same levels as the Control group, with the CDDP + flurbiprofen group

weighing less than CDDP group. The changes in the kidney weight indicated similar

12

tendencies as the body weight changes. However, this was attributed to the decreasing body

weight since the kidney weight per body weight was not altered significantly. Kidney injury

molecule 1 (Kim-1), kidney injury marker, was increased by CDDP. Notably, celecoxib

co-administration decreased and flurbiprofen co-application increased Kim-1 mRNA levels in

comparison to CDDP group (Fig. 5A and Fig. 5B). Accordingly, celecoxib attenuated and

flurbiprofen enhanced CIN in vivo, as well as in vitro. Also, the changes in mRNA levels of

Ho-1, Sod1, Nrf2, and Atg5 by CDDP are shown in Supplemental Table 2.

4. Discussion

We have examined the effects of 17 NSAIDs on CDDP-induced cytotoxicity in vitro

and revealed that several NSAIDs demonstrate the potential to attenuate or enhance cell injury

by CDDP. Notably, we observed that celecoxib significantly attenuated and flurbiprofen

markedly enhanced CIN. Initially, we predicted that the attenuating and enhancing effects of

NSAIDs on CIN were due to the differential inhibition selectivity between COX-1 and

COX-2, since celecoxib is a COX-2 selective inhibitor and flurbiprofen is highly selective for

COX-1. However, on assessing concomitant treatment with rofecoxib, with higher selectivity

for COX-2 than celecoxib, and ketorolac tromethamine, highly selective for COX-1 than

flurbiprofen, (Warner et al., 1999), they did not affect the results. Moreover, etodolac and

meloxicam, which like celecoxib are highly selective for COX-2, did not reduce

13

CDDP-induced cytotoxicity. Accordingly, the influence of NSAIDs is considered to be

independent of selectivity for COX inhibition. This observation cannot be explained without

the simultaneous evaluation of this study, as there is no previous report evaluating the

influence of NSAIDs on CDDP-induced cytotoxicity under similar conditions. Moreover,

previous reports utilized recombinant COX-1 and COX-2 to assess the inhibition of COX

activity by NSAIDs (Abdelall et al., 2016), and failed to evaluate the inhibition of COX

activity and calculate IC50 by NSAIDs in cellular COX-1 and COX-2. To further assess is

needed.

Antioxidant effects and autophagy are important approaches for attenuating CIN

(Kaushal and Shah, 2016; Rashtchizadeh et al., 2019). Cell dysfunction is induced by CDDP

via reactive oxygen species (ROS) production (Kawai et al., 2006). Therefore, eliminating

ROS and increasing resistance to oxidative stress by antioxidant substrates are valuable to

prevent CIN. In fact, the interaction between CDDP and food substances possessing

antioxidant effects has been investigated. For example, many researchers have assessed the

effect of curcumin in preventing CIN (Sahin et al., 2014; Trujillo et al., 2016). It has also been

reported that CIN is caused by mitochondrial dysfunction (Park et al., 2002), and autophagy is

known to remove the injured organelle. Decrease in the autophagy functions could induce

CIN (Takahashi et al., 2012). Therefore, it remains important to focus on the antioxidant and

autophagy functions. We evaluated whether N-acetylcysteine (NAC), which has a strong

14

antioxidant effect (Duval et al., 2019), attenuates CDDP-induced cytotoxicity. Co-addition of

NAC and CDDP showed higher IC50 than CDDP alone, similar to celecoxib. Hence, we

confirmed that the antioxidant effect is important to protect against cell dysfunction induced

by CDDP (Supplemental Fig. 2).

Celecoxib upregulated mRNA levels of antioxidant and autophagy markers in

comparison to CDDP alone. Flurbiprofen, however, did not downregulate these markers.

Therefore, it is speculated that CIN attenuation by celecoxib could be related to these markers.

The detailed mechanisms of CIN enhancement by flurbiprofen remain unclear. We have

previously reported that decreasing renal platinum accumulation attenuates CIN by regulating

the expression of renal transporters (Saito et al., 2017b). CDDP is mainly taken up by OCT2

and CTR1, and effluxed by MATE1 (Yokoo et al., 2009). We evaluated the platinum

accumulation in NRK-52E cells, and found that it was not changed by the co-addition of

celecoxib or flurbiprofen (data not shown). Therefore, these NSAIDs may not regulate renal

transporters, or the contribution of passive diffusion in the transport of CDDP may be

significant. In addition, CDDP is conjugated with glutathione followed by cysteine, and

CDDP-cysteine conjugation is suggested to be more toxic than CDDP (Townsend et al., 2009).

Therefore, NSAIDs may affect the metabolism of CDDP. xCT, a cystine/glutamate transporter,

is an important for the production of glutathione in cells (Thomas et al., 2015). Although we

evaluated xCT mRNA expression, xCT mRNA was not detected in the rat kidney cortex.

15

Hence, the involvement of the mechanisms of CDDP metabolism through xCT may be low.

However, the combination CDDP and flurbiprofen needs careful consideration. Flurbiprofen

axetil, flurbiprofen’s prodrug, is used in cancer pain in clinical settings (Wu et al., 2014). Thus,

patients administered CDDP may switch from flurbiprofen axetil to celecoxib, or alternative

medications such as acetaminophen and tramadol, and a detailed clinical evaluation needs to

be undertaken.

We assessed Sod1 and Atg5 mRNA levels to identify the in vivo mechanisms of CIN

attenuation by celecoxib. However, celecoxib co-administration did not increase these mRNA

levels (data not shown). Since the celecoxib mechanism in CIN might differ in vitro and in

vivo, we evaluated apoptosis to confirm this hypothesis. On assessing Bcl-2-associated X

protein/B-cell lymphoma-2 (Bax/Bcl-2) mRNA levels, as apoptosis markers, we observed that

these variations differed between the in vitro and in vivo evaluation in the CDDP + celecoxib

group (Supplemental Fig. 3). Therefore, the mechanism of actions were likely to be different.

We considered that the difference between the in vitro and in vivo mechanisms might be

associated with the NSAID activated gene 1 (NAG-1). NAG-1 is induced by NSAIDs,

regardless of COX activity, and is related to apoptosis (Zhang et al., 2014). Furthermore, it

has been suggested that celecoxib increases NAG-1 (Vaish et al., 2013). Therefore, regulating

Bax/Bcl-2 as an apoptosis marker might not be consistent owing to the differential effect of

CDDP and celecoxib on NAG-1 in vitro and in vivo. Furthermore, NSAIDs affect renal blood

16

flow (Herboczynska-Cedro and Vane, 1973). This change in renal blood flow may affect the

pharmacokinetics and the effect of NAG-1 in vitro and in vivo. Also, the selected time point

may be unsuitable for in vivo evaluation. CDDP accumulates in kidney for more than a week

(Saito et al., 2017c), whereas we evaluated the mRNA levels of several markers on day 5.

Therefore, the selected evaluation time point might not be appropriate, or some markers might

be regulated earlier than day 5. We evaluated the effects of celecoxib or flurbiprofen only on

Kim-1 mRNA expression. There was no change in the expression of Kim-1 compared with

Control (Supplemental Fig. 4). Therefore, the effects of co-administration of CDDP and

celecoxib or flurbiprofen may not be additive, but instead synergistic. Further investigation

into the underlying mechanisms needs to be conducted.

We evaluated serum creatinine level to confirm CIN by another renal injury marker.

Serum creatinine was increased by CDDP and celecoxib co-administration group was lower

serum creatinine level than CDDP group like Kim-1 mRNA expression (Supplemental Fig.

5A). However, flurbiprofen co-administration did not increase in comparison to CDDP alone.

This was thought to be due to differences in the sensitivity of kidney injury markers. As

Kim-1 was more sensitivity than serum creatinine (Vaidya et al., 2006), serum creatinine may

not reflect the effect of flurbiprofen co-administration. Moreover, we performed a

pathological evaluation of the rat kidney by using hematoxylin-eosin staining. As shown in

Supplemental Fig. 5B, there were no significant differences between all groups although it did

17

have strict quantification like TUNEL staining. It has been reported that low-dose CDDP does

not affect renal tissue and only increases renal injury markers (Yokoo et al., 2007). In addition,

increasing Kim-1 is earlier than pathological change (Han et al., 2002). As the dosage of

CDDP used in this study was close to a low-dose and 2–3 times higher than the clinical dose,

histological changes in the kidney may not occur and our results were consistent with a

previous report. In addition, we assessed the Timp-1 and Clusterin mRNA expression, which

are early kidney injury marker such as Kim-1 (Yao et al., 2007), and these markers were

increased by CDDP and suppressed by the co-administration of celecoxib (Supplemental Fig.

5C). Therefore, the effect of celecoxib may increase the protective action rather than

enhancing the tissue repair. However, this study was short-term and used a single injection of

CDDP; if it was long-term and used multiple injections, the effect of celecoxib on the repair

mechanisms may be recognized. Also, measuring urinary biomarkers may reveal new

strategies.

Flurbiprofen co-administration lost more weight than CDDP alone. Low weight due

to CDDP was thought to be caused by eating disorder. It has also been reported that CDDP

becomes more toxic by decreasing the concentration of magnesium in the body (Solanki et al.,

2014). For the above reasons, we considered that the increase in weight loss with the

co-administration of flurbiprofen was due to strong eating disorder or increased toxicity of

CDDP via enhancement of renal injury and accompanying hypomagnesemia. Also, we used

18

only male rats in this study because it has reported that OCT2 is more expressed in male and

CDDP is more distributed in the kidney (Yonezawa et al., 2005). Therefore, CIN is unlikely to

occur in female rats, but it may be caused similar to male rats by co-administration of

flurbiprofen if CDDP and flurbiprofen interact without transporters.

The concentrations of CDDP and NSAIDs used in the in vitro evaluation were almost

similar, or several-fold higher, compared to the blood concentration in clinical settings. As

these drugs are somewhat concentrated in the kidney, the screening results of the 17 NSAIDs

are highly relevant in clinical practice. However, it will be necessary to evaluate the actual

concentration of NSAIDs in the kidney and judge whether the in vivo dose matched the in

vitro concentration.

5. Conclusion

We observed that specific NSAIDs could either protect or worsen CDDP-induced

cytotoxicity. Among the 17 NSAIDs screened, celecoxib significantly alleviated cell injury by

CDDP in vitro via antioxidant effects and autophagy. Additionally, celecoxib significantly

attenuated CIN. Further clinical studies are crucial to evaluate the nephroprotective efficacy

of celecoxib in CIN.

Declaration of interest

19

None declared.

Acknowledgements

This work was supported by the Japan Society for the Promotion of Science (JSPS;

KAKENHI Grant Number JP19J11051, 19K16437 and 16H00487). We would like to thank

Editage (www.editage.jp) for English language editing.

References

Abdelall, E.K.A., Lamie, P.F., Ali, W.A.M., 2016. Cyclooxygenase-2 and 15-lipoxygenase

inhibition, synthesis, anti-inflammatory activity and ulcer liability of new celecoxib

analogues: Determination of region-specific pyrazole ring formation by NOESY.

Bioorg. Med. Chem. Lett. 26, 2893–2899. https://doi.org/10.1016/j.bmcl.2016.04.046.

de Jongh, F.E., van Veen, R.N., Veltman, S.J., de Wit, R., van der Burg, M.E., van den Bent,

M.J., Planting, A.S., Graveland, W.J., Stoter, G., Verweij, J., 2003. Weekly high-dose

cisplatin is a feasible treatment option: analysis on prognostic factors for toxicity in

400 patients. Br. J. Cancer 88, 1199–1206. https://doi.org/10.1038/sj.bjc.6600884.

Dobyan, D.C., Levi, J., Jacobs, C., Kosek, J., Weiner, M.W., 1980. Mechanism of

cis-platinum nephrotoxicity: II. Morphologic observations. J. Pharmacol. Exp. Ther.

213, 551–556.

Duval, A.P., Troquier, L., de Souza Silva, Q., Demartines, N., Dormond, O., 2019. Diclofenac

Potentiates Sorafenib-Based Treatments of Hepatocellular Carcinoma by Enhancing

Oxidative Stress. Cancers (Basel) 11, 1453–1466.

https://doi.org/10.3390/cancers11101453.

Fernández-Martínez, A.B., Benito Martínez, S., Lucio Cazaña, F.J., 2016. Intracellular

prostaglandin E2 mediates cisplatin-induced proximal tubular cell death. Biochim

Biophys Acta Mol Cell Res 1863, 293–302.

https://doi.org/10.1016/j.bbamcr.2015.11.035.

Han, W.K., Bailly, V., Abichandani, R., Thadhani, R., Bonventre, J.V., 2002. Kidney Injury

Molecule-1 (KIM-1): a novel biomarker for human renal proximal tubule injury.

Kidney Int. 62, 237–244. https://doi.org/10.1046/j.1523-1755.2002.00433.x.

Herboczynska-Cedro, K., Vane, J.R., 1973. Contribution of intrarenal generation of

20

prostaglandin to autoregulation of renal blood flow in the dog. Circ. Res. 33, 428–436.

https://doi.org/10.1161/01.res.33.4.428.

Honma, S., Takahashi, N., Shinohara, M., Nakamura, K., Mitazaki, S., Abe, S., Yoshida, M.,

2013. Amelioration of cisplatin-induced mouse renal lesions by a cyclooxygenase

(COX)-2 selective inhibitor. Eur. J. Pharmacol. 715, 181–188.

https://doi.org/10.1016/j.ejphar.2013.05.023.

Kaushal, G.P., Shah, S.V., 2016. Autophagy in acute kidney injury. Kidney Int. 89, 779–791.

https://doi.org/10.1016/j.kint.2015.11.021.

Kawai, Y., Nakao, T., Kunimura, N., Kohda, Y., Gemba, M., 2006. Relationship of

intracellular calcium and oxygen radicals to Cisplatin-related renal cell injury. J.

Pharmacol. Sci. 100, 65–72. https://doi.org/10.1254/jphs.fp0050661.

Kidera, Y., Kawakami, H., Sakiyama, T., Okamoto, K., Tanaka, K., Takeda, M., Kaneda, H.,

Nishina, S., Tsurutani, J., Fujiwara, K., Nomura, M., Yamazoe, Y., Chiba, Y., Nishida,

S., Tamura, T., Nakagawa, K., 2014. Risk factors for cisplatin-induced nephrotoxicity

and potential of magnesium supplementation for renal protection. PLoS ONE 9,

e101902. https://doi.org/10.1371/journal.pone.0101902.

Kobayashi, S., Kanai, S., 2013. Superoxide scavenging effects of some novel bis-ligands and

their solvated metal complexes prepared by the reaction of ligands with aluminum,

copper and lanthanum ions. Molecules 18, 6128–6141.

https://doi.org/10.3390/molecules18066128.

Loftin, C.D., Tiano, H.F., Langenbach, R., 2002. Phenotypes of the COX-deficient mice

indicate physiological and pathophysiological roles for COX-1 and COX-2.

Prostaglandins Other Lipid Mediat. 68–69, 177–185.

https://doi.org/10.1016/s0090-6980(02)00028-x.

Miller, R.P., Tadagavadi, R.K., Ramesh, G., Reeves, W.B., 2010. Mechanisms of Cisplatin

nephrotoxicity. Toxins (Basel) 2, 2490–2518. https://doi.org/10.3390/toxins2112490.

Pabla, N., Dong, Z., 2008. Cisplatin nephrotoxicity: mechanisms and renoprotective strategies.

Kidney Int. 73, 994–1007. https://doi.org/10.1038/sj.ki.5002786.

Park, M.S., De Leon, M., Devarajan, P., 2002. Cisplatin induces apoptosis in LLC-PK1 cells

via activation of mitochondrial pathways. J. Am. Soc. Nephrol. 13, 858–865.

Prestayko, A.W., D’Aoust, J.C., Issell, B.F., Crooke, S.T., 1979. Cisplatin

(cis-diamminedichloroplatinum II). Cancer Treat. Rev. 6, 17–39.

https://doi.org/10.1016/s0305-7372(79)80057-2.

Rashtchizadeh, N., Argani, H., Ghorbanihaghjo, A., Sanajou, D., Hosseini, V., Dastmalchi, S.,

Nazari Soltan Ahmad, S., 2019. AMPK: A promising molecular target for combating

cisplatin toxicities. Biochem. Pharmacol. 163, 94–100.

https://doi.org/10.1016/j.bcp.2019.02.006.

Rosenberg, B., VanCamp, L., Trosko, J.E., Mansour, V.H., 1969. Platinum compounds: a new

21

class of potent antitumour agents. Nature 222, 385–386.

https://doi.org/10.1038/222385a0.

Sahin, K., Orhan, C., Tuzcu, M., Muqbil, I., Sahin, N., Gencoglu, H., Guler, O., Padhye, S.B.,

Sarkar, F.H., Mohammad, R.M., 2014. Comparative in vivo evaluations of curcumin

and its analog difluorinated curcumin against cisplatin-induced nephrotoxicity. Biol.

Trace Elem. Res. 157, 156–163. https://doi.org/10.1007/s12011-014-9886-x.

Saito, Y., Kobayashi, M., Yamada, T., Kasashi, K., Honma, R., Takeuchi, S., Shimizu, Y.,

Kinoshita, I., Dosaka-Akita, H., Iseki, K., 2017a. Premedication with intravenous

magnesium has a protective effect against cisplatin-induced nephrotoxicity. Support

Care Cancer 25, 481–487. https://doi.org/10.1007/s00520-016-3426-5.

Saito, Y., Okamoto, K., Kobayashi, M., Narumi, K., Furugen, A., Yamada, T., Iseki, K., 2017b.

Magnesium co-administration decreases cisplatin-induced nephrotoxicity in the

multiple cisplatin administration. Life Sci. 189, 18–22.

https://doi.org/10.1016/j.lfs.2017.08.028.

Saito, Y., Okamoto, K., Kobayashi, M., Narumi, K., Yamada, T., Iseki, K., 2017c. Magnesium

attenuates cisplatin-induced nephrotoxicity by regulating the expression of renal

transporters. Eur. J. Pharmacol. 811, 191–198.

https://doi.org/10.1016/j.ejphar.2017.05.034.

Sato, K., Watanabe, S., Ohtsubo, A., Shoji, S., Ishikawa, D., Tanaka, T., Nozaki, K., Kondo,

R., Okajima, M., Miura, S., Tanaka, J., Sakagami, T., Koya, T., Kagamu, H.,

Yoshizawa, H., Narita, I., 2016. Nephrotoxicity of cisplatin combination

chemotherapy in thoracic malignancy patients with CKD risk factors. BMC Cancer 16,

222–227. https://doi.org/10.1186/s12885-016-2271-8.

Solanki, M.H., Chatterjee, P.K., Gupta, M., Xue, X., Plagov, A., Metz, M.H., Mintz, R.,

Singhal, P.C., Metz, C.N., 2014. Magnesium protects against cisplatin-induced acute

kidney injury by regulating platinum accumulation. Am. J. Physiol. Renal. Physiol.

307, F369–F384. https://doi.org/10.1152/ajprenal.00127.2014.

Takahashi, A., Kimura, T., Takabatake, Y., Namba, T., Kaimori, J., Kitamura, H., Matsui, I.,

Niimura, F., Matsusaka, T., Fujita, N., Yoshimori, T., Isaka, Y., Rakugi, H., 2012.

Autophagy guards against cisplatin-induced acute kidney injury. Am. J. Pathol. 180,

517–525. https://doi.org/10.1016/j.ajpath.2011.11.001.

Thomas, A.G., Sattler, R., Tendyke, K., Loiacono, K.A., Hansen, H., Sahni, V., Hashizume, Y.,

Rojas, C., Slusher, B.S., 2015. High-Throughput Assay Development for

Cystine-Glutamate Antiporter (xc-) Highlights Faster Cystine Uptake than Glutamate

Release in Glioma Cells. PLoS ONE 10, e0127785.

https://doi.org/10.1371/journal.pone.0127785.

Townsend, D.M., Tew, K.D., He, L., King, J.B., Hanigan, M.H., 2009. Role of glutathione

S-transferase Pi in cisplatin-induced nephrotoxicity. Biomed. Pharmacother. 63, 79–85.

22

https://doi.org/10.1016/j.biopha.2008.08.004.

Trujillo, J., Molina-Jijón, E., Medina-Campos, O.N., Rodríguez-Muñoz, R., Reyes, J.L.,

Loredo, M.L., Barrera-Oviedo, D., Pinzón, E., Rodríguez-Rangel, D.S.,

Pedraza-Chaverri, J., 2016. Curcumin prevents cisplatin-induced decrease in the tight

and adherens junctions: relation to oxidative stress. Food Funct. 7, 279–293.

https://doi.org/10.1039/c5fo00624d.

Tsuruya, K., Ninomiya, T., Tokumoto, M., Hirakawa, M., Masutani, K., Taniguchi, M.,

Fukuda, K., Kanai, H., Kishihara, K., Hirakata, H., Iida, M., 2003. Direct involvement

of the receptor-mediated apoptotic pathways in cisplatin-induced renal tubular cell

death. Kidney Int. 63, 72–82. https://doi.org/10.1046/j.1523-1755.2003.00709.x.

Vaidya, V.S., Ramirez, V., Ichimura, T., Bobadilla, N.A., Bonventre, J.V., 2006. Urinary

kidney injury molecule-1: a sensitive quantitative biomarker for early detection of

kidney tubular injury. Am. J. Physiol. Renal Physiol. 290, 517–529.

https://doi.org/10.1152/ajprenal.00291.2005.

Vaish, V., Piplani, H., Rana, C., Vaiphei, K., Sanyal, S.N., 2013. NSAIDs may regulate

EGR-1-mediated induction of reactive oxygen species and non-steroidal

anti-inflammatory drug-induced gene (NAG)-1 to initiate intrinsic pathway of

apoptosis for the chemoprevention of colorectal cancer. Mol. Cell. Biochem. 378,

47–64. https://doi.org/10.1007/s11010-013-1593-y.

Vane, J.R., 1971. Inhibition of prostaglandin synthesis as a mechanism of action for

aspirin-like drugs. Nature New Biol. 231, 232–235.

https://doi.org/10.1038/newbio231232a0.

Warner, T.D., Giuliano, F., Vojnovic, I., Bukasa, A., Mitchell, J.A., Vane, J.R., 1999.

Nonsteroid drug selectivities for cyclo-oxygenase-1 rather than cyclo-oxygenase-2 are

associated with human gastrointestinal toxicity: a full in vitro analysis. Proc. Natl.

Acad. Sci. U.S.A. 96, 7563–7568. https://doi.org/10.1073/pnas.96.13.7563.

World Health Organization, 1996. Cancer pain relief: with a guide to opioid availability,

second ed. World Health Organization, Geneva.

Wu, T.T., Wang, Z.G., Ou, W.L., Wang, J., Yao, G.Q., Yang, B., Rao, Z.G., Gao, J.F., Zhang,

B.C., 2014. Intravenous flurbiprofen axetil enhances analgesic effect of opioids in

patients with refractory cancer pain by increasing plasma β-endorphin. Asian Pac. J.

Cancer Prev. 15, 10855–10860. https://doi.org/10.7314/apjcp.2014.15.24.10855.

Yao, X., Panichpisal, K., Kurtzman, N., Nugent, K., 2007. Cisplatin nephrotoxicity: a review.

Am. J. Med. Sci. 334, 115–124. https://doi.org/10.1097/MAJ.0b013e31812dfe1e.

Yokoo, K., Murakami, R., Matsuzaki, T., Yoshitome, K., Hamada, A., Saito, H., 2009.

Enhanced renal accumulation of cisplatin via renal organic cation transporter

deteriorates acute kidney injury in hypomagnesemic rats. Clin. Exp. Nephrol. 13,

578–584. https://doi.org/10.1007/s10157-009-0215-1.

23

Yokoo, S., Yonezawa, A., Masuda, S., Fukatsu, A., Katsura, T., Inui, K., 2007. Differential

contribution of organic cation transporters, OCT2 and MATE1, in platinum

agent-induced nephrotoxicity. Biochem. Pharmacol. 74, 477–487.

https://doi.org/10.1016/j.bcp.2007.03.004.

Yonezawa, A., Masuda, S., Nishihara, K., Yano, I., Katsura, T., Inui, K., 2005. Association

between tubular toxicity of cisplatin and expression of organic cation transporter

rOCT2 (Slc22a2) in the rat. Biochem. Pharmacol. 70, 1823–1831.

https://doi.org/10.1016/j.bcp.2005.09.020.

Yoshida, T., Niho, S., Toda, M., Goto, K., Yoh, K., Umemura, S., Matsumoto, S., Ohmatsu, H.,

Ohe, Y., 2014. Protective effect of magnesium preloading on cisplatin-induced

nephrotoxicity: a retrospective study. Jpn. J. Clin. Oncol. 44, 346–354.

https://doi.org/10.1093/jjco/hyu004.

Zhang, Z., Wu, L., Wang, J., Li, G., Feng, D., Zhang, B., Li, L., Yang, J., Ma, L., Qin, H.,

2014. Opposing effects of PI3K/Akt and Smad-Dependent signaling pathways in

NAG-1-induced glioblastoma cell apoptosis. PLoS ONE 9, e96283.

https://doi.org/10.1371/journal.pone.0096283.

Figure Legends

Fig. 1: Transporter expression on NRK-52E cells and the characteristics of NRK-52E

cells treated with CDDP. (A) Upper picture is the apical side and lower picture is the

basolateral side of renal transporters by RT-PCR. L means left sample used rat kidney cortex’s

cDNA and R means right sample used NRK-52E cells’ cDNA. (B) Cells were incubated with

CDDP (0–30 μM). (C) Clusterin mRNA was normalized to Actin. Cells in the CDDP group

were incubated with CDDP (10 μM) for 48 h. The expression level of the Control group was

arbitrarily set at 1.0. **P < 0.01 compared with the Control group; an unpaired Student’s t-test.

Data are presented as means with S.E.M., for three independent experiments.

Fig. 2: Effect of NSAIDs on CDDP-induced cytotoxicity and Clusterin mRNA levels in

24

CDDP-treated NRK-52E cells. (A) The values were the ratio of IC50 of co-addition of CDDP

and NSAIDs to IC50 of CDDP alone. The concentration of NSAIDs co-addition with CDDP

was defined as the concentration at which no effect on cell viability was observed when

treated with NSAIDs only (described Table 1). Clusterin mRNA levels were altered following

incubation with CDDP (10 μM) and (B) celecoxib (50 μM) or (C) flurbiprofen (200 μM) for

48 h. Clusterin mRNA was normalized to Actin. The expression level of the Control group

was arbitrarily set at 1.0. *P < 0.05, **P < 0.01 and ***P < 0.001 compared with the Control

group, ††P < 0.01 and †††P < 0.001 compared with the CDDP group, ‡‡P < 0.01 and ‡‡‡P <

0.001 compared with the celecoxib or flurbiprofen group; Tukey’s post-hoc test. Data are

presented as means with S.E.M. for three independent experiments.

Fig. 3: Effect of NSAIDs on mRNA levels of antioxidant markers in CDDP-treated

NRK-52E cells. (A) Ho-1, (B) Sod1 and (C) Nrf2 mRNA were normalized to Actin. Cells

were incubated with CDDP (10 μM) and each NSAID (concentration is shown in Table 1) for

48 h. The expression level of the CDDP group was arbitrarily set at 1.0. †P < 0.05, ††P < 0.01

and †††P < 0.001 compared with the CDDP group; Dunnett’s test. Data are presented as means

with S.E.M. for three independent experiments.

Fig. 4: Effect of NSAIDs on mRNA levels of autophagy markers in CDDP-treated

NRK-52E cells. (A) Atg5, Atg7, Lc 3 and Beclin 1 mRNA levels were altered by incubation

with CDDP (10 μM), celecoxib (50 μM) and flurbiprofen (200 μM) for 48 h. The mRNAs

25

were normalized to Actin. The expression level of the Control group was arbitrarily set at 1.0.

**P < 0.01 and ***P < 0.001 compared with the Control group, †P < 0.05 and †††P < 0.001

compared with the CDDP group, ‡P < 0.05, ‡‡P < 0.01 and ‡‡‡P < 0.001 compared with the

CDDP + celecoxib group; Tukey’s post-hoc test. (B) Atg5 mRNA was normalized to Actin.

Cells were incubated with CDDP (10 μM) and each NSAID (concentration is shown in Table

1) for 48 h. The expression level of the CDDP group was arbitrarily set at 1.0. †††P < 0.001

compared with the CDDP group; Dunnett’s test. Data are presented as means with S.E.M. for

three independent experiments.

Fig. 5: Effect of celecoxib and flurbiprofen on rats treated with CDDP. (A) Kim-1 and

Actin mRNA expression by RT-PCR, and (B) mRNA was quantified by ImageJ analysis

software. Kim-1 was normalized to Actin. The expression level of the Control group was

arbitrarily set at 1.0. **P < 0.01 and ***P < 0.001 compared with the Control group, †P < 0.05

and ††P < 0.01 compared with the CDDP group, ‡‡‡P < 0.001 compared with the CDDP +

celecoxib group; Tukey’s post-hoc test. Data are presented as means with S.D., n = 5–7 per

group.

1

Table 1. MTT assay of NSAIDs alone and setting of concentration for co-addition of CDDP

NSAIDs IC50 (μM) Concentration of co-addition (μM)

Aspirin > 200 200

Celecoxib 63.4 ± 1.5 50

Diclofenac sodium > 200 200

Etodolac > 200 200

Flufenamic acid > 200 200

Flurbiprofen > 200 200

Ibuprofen > 200 200

Indomethacin > 200 25

Ketorolac tromethamine > 200 200

Lornoxicam > 10 10

Mefenamic acid > 200 200

Meloxicam > 80 80

Naproxen > 10 10

Oxaprozin > 200 200

Piroxicam > 200 200

Rofecoxib > 25 25

Zaltoprofen > 200 200

Cells were incubated with each NSAID. MTT assay was evaluated until each saturation concentration was established. Concentration of the

NSAID co-added with CDDP was defined as no effect on NRK-52E cells within the range of 95–120% cell viability when treated with NSAIDs

2

alone. Data are presented as means ± S.E.M., of three independent experiments.

1

Table 2. Correlations between cell viability and antioxidant or autophagy markers

R P value

IC50 of MTT assay Ho-1 mRNA level 0.505 0.0385c

Sod1 mRNA level 0.758 < 0.001a

Nrf2 mRNA level 0.685 0.00242b

Atg5 mRNA level 0.898 < 0.001a

IC50 of MTT assay is shown in Fig. 2A, and Ho-1, Sod1, Nrf2 and Atg5 mRNA level are shown in Fig. 3A, Fig. 3B, Fig. 3C and Fig. 4B

respectively. R means correlation coefficient.

a P < 0.001 compared between IC50 of MTT assay and each mRNA level; Pearson correlation coefficient.

b P < 0.01 compared between IC50 of MTT assay and each mRNA level.

c P < 0.05 compared between IC50 of MTT assay and each mRNA level.

1

Table 3. Variation of body and kidney weight in rats

Body weight (g) Kidney weight (g) Kidney weight/body weight

Day 1 (baseline) Day 5 Day 5 – Day 1 Day 5 Day 5

Control 193.9 ± 4.7 219.4 ± 4.5 25.5 ± 1.3 2.08 ± 0.09 0.0095 ± 0.0005

CDDP 191.3 ± 8.8 184.4 ± 8.5 -7.0 ± 4.6a 1.75 ± 0.17b 0.0095 ± 0.0008

CDDP + celecoxib 194.9 ± 5.7 221.0 ± 5.3 26.2 ± 2.2c 1.99 ± 0.09 0.0090 ± 0.0003

CDDP + flurbiprofen 191.8 ± 10.8 162.0 ± 17.6 -29.8 ± 15.5a, d, e 1.62 ± 0.21a, f 0.0100 ± 0.0008

Body weight at baseline and day 5 was measured. Variation of body weight from baseline was compared between groups. Kidney weight was

total right and left kidney. Data are presented as means ± S.D., n = 5–7 per group.

a P < 0.001 compared with Control group; Tukey’s post-hoc test.

b P < 0.01 compared with Control group.

c P < 0.001 compared with CDDP group.

d P < 0.01 compared with CDDP group.

e P < 0.001 compared with CDDP + celecoxib group.

f P < 0.01 compared with CDDP + celecoxib group.